A. Adediji, A.M. Tukur and K.A. Adepoju. A. Adediji, Department of Geography, Obafemi Awolowo University, Ile-Ife, Nigeria

Transcription

1 Iranica Journal of Energy & Environment 1 (3): , 2010 ISSN IJEE an Official Peer Reviewed Journal of Babol Noshirvani University of Technology Assessment of Revised Universal Soil Loss Equation (RUSLE) in Katsina Area, Katsina State of Nigeria using Remote Sensing (RS) and Geographic Information System (GIS) A. Adediji, A.M. Tukur and K.A. Adepoju 1 Department of Geography, Obafemi Awolowo University, Ile-Ife, Nigeria 2 Regional Centre for Space Science and Technology Education, OAU, Ile-Ife, Nigeria BUT Abstract: The Revised Universal Soil Loss Equation (RUSLE) parameters were assessed using Satellite Remote Sensing (RS) and GIS with a view to model soil erosion in Katsina area of Katsina State of Nigeria. Data on parameters such as slope factors, crop cover and management practice support (P) were obtained from obtained for Katsina area for Digital Elevation Model (DEM) and Landsat ETM +, 2002 of the area. The estimated potential mean annual soil loss of ton/ac/yr based on the refined RUSLE was obtained for the study area. Also, the potential erosion rates from the erosion classes identified ranged from 0.0 to ton/ac/yr. About 65.47% of the study area was classified under the first class with erosion rate between 0.0 and 10 ton/ac/yr. The most severely eroded area with rates of erosion between and 4, ton/ac/yr accounted for about 1.86% of the study area. On the whole, this study has demonstrated the significance of Satellite (RS) and GIS technologies in modeling erosion. Key words: Assessment % RUSLE % Satellite Imagery and Potential Soil Erosion INTRODUCTION A number of parametric models for predicting soil erosion exist e.g. Universal Soil Loss Equating (USLE), Erosion can be described as the wearing away Revised USLE (RUSLE), Modified USLE of the earths surface material by wind, water, ice (MUSLE),Water Erosion Prediction Project (WEPP) and or gravity. Problems associated with the accelerated Soil Loss Estimation Model for Southern African erosion persisted for more than a million geologic (SLEMSA). Both USLE and WEPP by [1], are widely years ago in almost all parts of the globe. However, used in North American and even adopted and applied the situation is compounded in recent times by in other regions of the world (e.g. [2] in Southwestern man s increasing interactions with the environment Nigeria; [3] in South eastern Nigeria and [4] in Milewa and the fact that data collections on soil erosion is Catchments, Kenya). For instance, [2] in South western usually capital intensive as well as a time consuming Nigeria substituted the Wischmcier and South s R exercise. Hence, global extrapolation of a few data factor with the [5] rainfall erosivity (R) index which collected through various diverse and non- was not adequately modeled by USLE because of standardized methods often leads to gross error high intensity of tropical rainfalls. [6] Examined the and consequently, it can lead to wrong assessment on relative efficiencies of erosivity indices in the soil loss critical policy issues. In this regard, remote sensing equation in Southeastern Nigeria. especially Satellite Remote Sensing provides a Also, some published studies existed on the convenient technique to solve this problem. Remote application of RS and GIS technologies to modeling of soil Sensing (RS) and Geographic Information System erosion in other parts of the worlds (e.g. [7], [8] and [4]). (GIS) enable manipulation of spatial data of various However, there is little or no known work on the types. The ability to extract overlay and delineate any application of GIS to erosion modeling in Northern Nigeria land characteristics make GIS suitable for soil erosion if not in the whole of Nigeria. Thus, the more recent modeling. version of the USLE, the Revised Universal Soil applied in Corresponding Author: A. Adediji, Department of Geography, Obafemi Awolowo University, Ile-Ife, Nigeria 255

2 Fig. 1: Map of Katsina State showing the study area this study to erosion modeling in Katsina area of Katsina Study Objective: State of Nigeria. RUSLE uses the same empirical principles The specific objectives of this study are to: as the USLE but includes improved rainfall erosivity factor, incorporation of the influence of profile convexity / C Identify areas that have been affected by soil erosion concavity using segmentation of irregular slopes and C Estimate the potential soil loss from the affected area improved empirical equation for computing slope factor using RUSLE (LS). C Generate Erosion Hazard Map for the study area. The RUSLE can be expressed as:- Study Area: Katsina area in Katsina State of Nigeria constitute the study area. It lies between Latitude N A = RXKXLSXCXP (1) and 13 N and Longitude 7 30 E and 8 East of Greenwich 2 Meridian (Fig. 1). It covers an area of about 3,025km. The Where A = average soil loss (ton/ac/yr); R = rainfall area is bordered to the south by Musawa and to the north erosivity factor (MJ.mm / ha.hr.yr); soil erodibility factor by Dankama. The area is further bordered to the West by (ton/ac/unit R). LS = slope factor (dimensionless); C = Ruma and to the East by Kazuare (Jigawa State). The area cover factor (dimensionless) and P = prevention practices also covers Rimi, Kanya, Charanchi and Batagarawa. factor (dimensionless). The effective determination of the Local Government Areas as well as Bindawa, Mane and RUSLE factor is fundamental to estimation of soil loss Mashi. The area is drained by major rivers such as the from cropland and rangeland. Specifically, interfacing GIS Koza, Sabke, Tagwai and Gada System in the northern analysis capabilities with the RUSLE provides the part of the state. resource specialist with a tool to visualize quickly the soil The area is characterized by Tropical Continental erosion potential area based on several major Climate (Sudan Type). The mean annual rainfall ranges environmental parameters for large areas. Therefore, the between 600 and 1100mm. The temperature during eroded area in Katsina area will be categorized into Harmattan season ranges from 18 to 27 C. Also, the various classes using RS and GIS techniques. maximum temperature ranges between 29 and 38 C. 256

3 Most parts of the study area are underlain by light sandy These values are inputted into equation 2 to derived soils of low-medium fertility. The vegetation of the area R factor for the study area (Table 1). consists of trees characterized by long tap roots and thick barks (e.g. Acacia sp. and Eucalyptus sp.). This makes it Slope Factor (LS): Derivation of slope factor (LS) possible for them to withstand the long dry season and for the area involved generation of Digital Elevation bush fire. Specially, Katsina area plays a leading role in Model (DEM) for the study area from the topographical the production of a lot of cash and food crops (e.g. sheet. The contour data were extracted from a topo Gossypium sp. Arachis hupogaea, Phaseoulus sp., sheet (1:100,000) of the study area through scanning Penisetum americanum and Oryza sativa) for the and manual digitizing using Arc Map. The DEM country. generated was converted to a raster file with The dramatic population growth, overgrazing and resolution set at 30m.[10] Slope layer was derived from large dependency on agriculture as well as the use of the DEM. The LS factor was determined using equation wood as fuel are responsible for land use /cover dynamics developed. [11] in the area. Land degradation is quite common in the area and in fact, large area of the land surface is under the m n LS = (As/22.13) (Sin 2 /0.09) (3) threat of desertification. This consequently exposes the soil to erosion. Where As = upslope contributing areas per unit width of cell spacing; 2 =slope angle (degrees), m and n are Study Method: Data used for evaluation of RUSLE factors exponent of slope parameters for slope length and and generation of Erosion Hazard Map in this study were gradient and the typical values of m and n are and obtained from Secondary sources. These data were , respectively. Lower values of m = 0.4 and n =1.1 processed using the maximum likelihood classification were used for the study area because of the undistributed algorithm in ERDAS Imagine and the spatial analyst and nature of the area. 3D analyst extensions of ARC GIS 9.2 software for spatial analysis. Land Cover Factor (C): The classification of a Landsat The materials used include topographic sheets, ETM + image was done using ERDAS IMAGINE software Landsat ETM+2002 (resolution of 30m and path /Row of which was used to prepare land use/ land cover map of P189R51) obtained from the Global Land Cover facility the study area. The result of the classification was used (GLCF) website, rainfall distribution and soil erodibility to derive the C-factor for each of the land cover identified shape file were collected from African Regional Centre for (Table 2). Space Science and Technology Education (ARCSSTEE), OAU,Ile-Ife, Nigeria. Rainfall Erosivity Factor (R): The average annual rainfall data of the study area for the period of 25 years ( ) was obtained from the Nigeria Metrological Agency (NIMET). Although the annual R index is not directly linked to annual rainfall, however, [9] in West Africa has shown that: The main annual rainfall erosivity over 10years = mean annual rainfall *a.(2) Where a = 0.05 in most cases ± 0.05 = 0.6 near the sea (<40km) = 0.3 to 0.2 in tropical mountain areas = 0.1 in Mediterranean mountain area The rainfall pattern of the study area is such that rainfall increases southward with the northern parts receiving an average of 600mm of rainfall annually while the southern parts experience an average of 1000mm annually. Table 1: C factor and Areal Coverage for Classes Derived from LandSat Image Land-Use/Cover Classes Areal Coverage(Acres) C Factor Town 23, Degraded Forests Savannah/Grassland 323, Agriculture (sparse) 362, Water bodies 15, Bare land 17, Table 2: K values for different soil textures Organic matter content (%) Textural class Fine sand Very fine sand Loamy sand Loamy Very fine sand Sandy loam Very fine sandy loam Silt loam Clay loam Silty clay loam Clay

4 Data Collection Topo Map Rainfall Map Soil Map Landsat Image DEM Vector Conversion Vector Conversion Image Classification Slope Rainfall Grid Soil Grid Landuse/cover Grid LS Factor R Factor K Factor C Factor P Factor Multiplication Soil Soil Erosion Erosion Hazard Hazard Map Map Fig. 2: Flow Chart of Study Methodology Soil Erodibility Factor (k): A descriptive soil map of the study area was obtained from previous Food and Agricultural Organization (FAO) survey in Nigeria. The description in the Attribute Table enabled us to compute k factor for each textural class using the values in Tables 3. Practice Management Factor (P): The P factor map was prepared from land use/cover map. [11] method was employed in this study to assign the P values for the study area. Table 3: R factor Values Average Annual Rainfall 600mm mm 450 R factor Determination of Potential Annual Soil Loss: The estimated annual soil loss was computed by multiplying the grid layers of the factors described above in the GIS environment using Arc Map. The summary of the study method employed in this study is shown in Figure

5 RESULTS AND DISCUSSION The erosion values obtained through RUSLE can vary considerably due to varying weather conditions. The results obtained for the RUSLE factors are shown in Table 2-4 and Figure Rainfall Erosivity Factor (R): The distribution of the average annual rainfall distribution of Kastina over 22years period is shown in Fig.3. It is quite evident from Fig.3 that rainfall in the study area increases southwards with the northern part experiencing an average of 600mm of rainfall annually while the southern parts receive a mean of 1000mm annually. The map of rainfall erosivity index (R) derived for the study area is shown in Fig.4. The R factor values obtained (see Fig.4) compared favourably Table 4: P Factor Values (Wischmeier and Smith, 1978 (MCMLXXVIII)) Land Use Type Slope % P-Factor Agriculture Other Land All 1.00 with those obtained by [4] in Malewa Catchments, Kenya. Although, [4] employed the erosivity regression equation proposed by [12] for Kenya hinterland, this R factor values of between and N/h for areas with average annual rainfall values between 600 and 800mm compared closely with values obtained in this study. Fig. 3: Annual Rainfall Distribution in the Study Area Fig. 4: Rainfall Erosivity, R, Map of the Study Area 259

6 Fig. 5: Land-Cover/Land-Use Classes of the Study Area Fig. 6: Soil Erodibility, K factor Map of the Study area 260

7 (m) Iranica J. Energy & Environ., 1 (3): , 2010 Fig. 7: DEM of the Study Area Fig. 8: Slope factor Map 261

8 Fig. 9: Support Practices Factor, P, Map of the Study Area Fig. 10: Erosion Hazard Map 262

9 Fig. 11: Histogram of soil loss Soil Erodibility Factor (k): Soil erodibility (k) represents both susceptibility of soil to erosion and the rate of runoff as measured under the standard and plot condition. The erodibility index map derived from FAO s Soil map of the study area is shown in Fig.5. The k-values obtained for the study area ranged from 0.04 to land. The C-factor map for the study area is shown in Fig. 8. The result on management / support practice (P) is shown in Fig 9. The support practice factor affects erosion by modifying the flow pattern or direction of surface runoff as well as reducing the amount and rate of runoff. The Slope/Topographical Factor (SL): The slope/ RUSLE Model: The mean annual soil loss estimated for topographical factor (SL) depends on both the length the study area using RUSLE was put at ton/ac/yr. and gradient of slope. It has been observed that soil Based on the model, the study area was classified into ten loss increases more rapidly with slope steepness than erosion classes ranging from 0.0 to ton/ac/yr it does with slope length. This study employed the (Fig. 10). However, 65.47% of the Kastina area is within modified SL equation developed by [11] instead of the first class with erosion rates ranged from 0.0 to 10 ton/ Wischmeier and Smith s SL to determine the upslope ac/yr. This is considered to be within the moderate range ton/ac/yr, respectively. The two classes The most severe eroded areas with erosion rates of combined cover 32.65% of the area shown in Fig. 11. between and 4, ton/ac/yr accounts for 1.86% It is evident from in depth analysis of all factors of the study area. The areas within the third and fourth that SL factor seems to have a significant effect on the class have severe erosion rates of between and estimated total soil loss in the area. This is because the and and ton/ac/yr, respectively. The areas mostly affected by erosion within the study area two classes combined cover 32.65% of the area shown in coincided with the areas where SL factor is the highest Fig. 11. contributing area. The DEM and SL factor map generated It is evident from in depth analysis of all factors that for the study area are shown in Fig. 6 and 7. SL factor seems to have significant effect on the estimated total soil loss in the area. This is because the areas mostly Crop Cover (L) and Management Practice (P) factor: As affected by erosion within the study area coincided with evident from Table 4 and Fig. 8 (histogram) of land the areas where SL factor is highest. use/cover distribution, the greater proportion of the study area is under farmland (including grazing) and savanna. CONCLUSION Also, towns /built-up account for small proportion of the area (Fig. 8). The C factor for land use /cover classified in A detail evaluation of RUSLE parameters in Katsina the area ranges from 0.02 (degraded forest) to 0.16 for area of Katsina State, Nigeria was carried out in this study agricultural land. This implies that erosion is lower under using Remote Sensing (RS) and GIS platforms. Ten the degraded vegetated area than the agricultural/grazing erosion classes were classified from the Landsat ETM + 263

10 image of the area. The landuse/cover classes identified are 5. Morgan, R.P.C., Field studies of Rain Splash builtup/towns degraded forests, farmlands, grassland, Erosion Earth Surface Processes, 3: bare land,and water bodies. Farmlands (36.7%) followed 6. Chinatu, T.N. and U. Charles, Relative by -grasslands (32.6%) covered the largest extent of the Efficiencies of Erosivity Indices in Soil Loss study area overall. The potential rates of soil loss from the Prediction in Southeastern Nigeria. J. Engineering ten classes ranged from 0.0 to ton/ac/yr. However, and Appl. Sci., (6): the mean soil loss estimated for the study area was put at 7. El-Swaify, S.A. and E.W. Danger, ton/ac/yr. The most severely eroded part of the Erodibilities of selected tropical soils in relation to study area account for about 1.86% of the total area structural and hydrologic parameter. In; Foster, G.R. extent. On the whole, this study has demonstrated (ed), Soil Erosion Prediction and control. Soil and conclusively that Remote Sensing and GIS are useful Water conservation society, Ankery, pp: tools for modeling soil erosion, evaluating various 8. Hesadi, H., K.H. Jalili and M. Hadidi, disturbance alternative and spatial optimization of Applying RS and GIS to Soil Erosion and Sediment conservation measures. Estimation by PSTAC Model. A case study of Kenesht Watershed in Kermanshah, Iran. REFERENCES gisdevelopment.net 9. Roose, E., Land Husbandry-Components 1. Wischmeier, W.H. and D.D. Smith, and strategy. FAO Corporate Document Repository, Predicting rainfall erosion losses, a guide to 70 FAO Soil Bulletin ISBN Chapter 5. conservation planning. Agriculture, Washington Lttp:// D.C., pp: Tukur, A.M., Modeling of Soil Erosion in 2. Jeje, L.K., O.O. Ogunkoya and A. Adediji, Kastina Area, Kastina State, Nigeria using RS and Adapting the Universal Soil Loss Equation. GIS Unpublished PGD Dissertation, ARCSSTEE, USLE to Southeastern Nigeria. Nigerian J. Sci., Ile-Ife, Nigeria, pp: (NJS), 31: Moore, I.D. and J.P. Wilson, Length-Slope 3. Igwe, C.A., F.O.R. Akanigbo and J.S.C. Mbagwu, factors for the Revised Universal Soil Loss Equation: Application of SLEMSA and USLE Erosion Simplified Methods of Estimation. J. Soil Sci. Water Models for Potential Erosion Hazard Mapping in Conservation, 45(5): Southeastern Nigeria, Int. Agrophysics, 13: Breches, B., Aggregate stability as an 4. Odongo, V.O., Evaluating the Topographic indicator of soil susceptibility to runoff and factor in the Universal Soil Loss Equation Using GIS erosion, validation at several levels, In: Catena, Techniques for Malewa Catchment, Nawasha, 47(2): Kenya. PGD Thesis, ARCSSTEE, Ile-Ife, Nigeri, pp: